Action to Cure Kidney Cancer

Test would be first to detect deadly kidney cancer

Paul CairnsOn March 2, 2006, Action to Cure Kidney Cancer (ACKC) awarded a $45,000 grant to Paul Cairns, Ph.D., of Fox Chase Cancer Center in Philadelphia to purchase an important piece of equipment to help his laboratory develop a urine test that will detect all types of kidney cancer (renal cell carcinoma) in the earliest stages of disease. The new equipment will speed up the project and result in a more accurate test--which would be the first to be used against this silent killer.

Dr. Cairns' work is part of the Early Detection Research Network, a program set up by the National Cancer Institute that issues grants designed to accelerate collaboration and speed up the translation of early detection projects to clinical settings.

For kidney cancer, early detection is especially critical. When the disease is confined to the kidney, surgery--either partial or complete removal of one kidney--enables many patients to live healthy lives for years. Because there is currently no effective method to screen for kidney cancer, and because the disease often presents no symptoms, more than 30% of patients are diagnosed after they have advanced disease. In its later stages, kidney cancer metastasizes-- or spreads--to other organs, such as the liver, lungs, bone, or brain. Late stage patients have extremely limited treatment options and fewer than 10% survive five years.

This new piece of equipment--known as a real-time PCR (polymerase chain reaction) detection system and manufactured by Applied Biosystems--allows Dr. Cairns' lab to further develop their urine screen for kidney cancer. The methodology is called quantitative real-time methylation-specific PCR, and it detects changes in patterns of gene methylation, one of the processes that can "hit" a tumor suppressor gene --turning it "off." Gene changes are the trigger that causes a normal cell to turn into cancer. In the case of the urine screen, the trial panel of six genes that they are currently working with can detect kidney cancer in almost 90% of patients, with no false positives. Dr. Cairns' lab is now looking to detect gene changes in individuals with early stage kidney tumors, specifically ones that are less than 3 centimeters in diameter.

How it works

To develop the screen, Dr. Cairns' lab compared tumor with normal cells to identify target genes that were methylated in kidney cancer. Now, the task is to increase the number of genes in the panel in order to increase the sensitivity of the test, aiming toward 100% predictive value.

"For the urine screen--we use it (the new equipment) to analyze the methylation status of genes from kidney tumor cells, and then we use it to see if we can detect the same methylation in urine DNA of the same patients," says Dr. Cairns.

It sounds much simpler than it actually is. The job is a massive puzzle that requires meticulous work to be performed repeatedly. The gene changes must be absent from the urine of healthy volunteers. The panel of genes must be representative of the changes that occur in all types of kidney cancer, from clear cell (which represents about 80% of kidney cancer cases) to papillary, chromophobe, and other rarer types.

"Kidney tumors, like all other cancers, are extremely heterogeneous in their genetics and biology," says Dr. Cairns. "That's why people with similar size, grade, and stage of tumors can have diseases that behave completely differently."

By the time the project is complete, hundreds of urine samples will have been screened, each many times. The ABI system will both speed up the work and increase the accuracy of the test. Because the equipment works in real time, the lab can work more quickly: "We can read it off the screen in real time--we don't have to do a gel analysis afterwards."

The power of the machine also allows the test to be more accurate in two ways: sensitivity and specificity. The sensitivity is nearly ten times that of the technology previously used by Dr. Cairns' lab--able to detect one cancer cell among 10,000 cells. Therefore the resulting test will have fewer false negatives--fewer patients will think they are disease free when they are not. A test with more specificity is less likely to produce false positives, or patients who think they have cancer but are then shown to be free from disease.

"I would say that kidney cancer is an understudied disease." Dr. Cairns says that it is notable how little we know about kidney cancer compared to some other cancers. "One of the main areas I'm interested in is that we have absolutely no idea which are the key genes that drive kidney cancer. We know VHL [von Hippel Lindau], which is the predisposition gene, and we know it is almost certainly the first and key gene that is hit in the majority--but not all--cases of sporadic [non-hereditary] clear cell cancer. But when you model cancer by age and incidence, you can predict that there may need to be six or seven genes mutated in kidney cancer--and we have no idea what five or six of those are."

The future

The ABI system purchased by ACKC provides quantitative data, which could open a window for Dr. Cairns into learning more about the behavior and natural history of kidney cancer.

"A quantitative assay really allows you to think about extending the use of methylation not just to early detection, but to prediction, prognosis, and monitoring," he says. "Instead of just telling you whether a gene is methylated or not, it tells you the quantitative level of methylation." Comparing methylation levels may help doctors confirm whether surgery has cured a patient or help detect recurrence early.

Dr. Cairns is interested in studying the gene profiles of "matched" pairs of tumors from patients with different outcomes to see if methylation can be used to predict whether a patient's cancer will be indolent or more active. Once identified, new genes could be added to the screening panel.

Methylation might have other applications for kidney cancer patients. "Perhaps most interestingly, you might be able to use this to truly monitor a patient that's receiving therapy," Dr. Cairns says. "If the level of methylation dropped during chemotherapy, you could argue you've got a Œsurrogate marker' of response." Similar technologies are in advanced testing for some other cancers.

"Personalized or individualized treatment is a very hot area at the moment," says Dr. Cairns. "I very much would like to work with tumors from patients who respond better and those who don't respond so well to promising drugs for kidney cancer. Seeing which genes are methylated will help me learn more about the biology of how the tumors respond to these drugs." Dr. Cairns says these genes could also be used to predict whether a patient might benefit from a particular treatment.

Dr. Cairns' urine screen is still in feasibility testing. The next stage is to perform larger validation studies to prove the screen is effective and clinically valuable, and then it would move on to clinical trials to demonstrate its use in larger populations.

Challenges

While Dr. Cairns finds this an exciting time in cancer research, he can also see that patients might be frustrated with the slow progress of the field.

"Maybe one point that doesn't get across to patients is just how stunningly complex cancer is as a disease, compared to other diseases," he says. There are 140 different cancers, and each patient's cancer can behave differently. Advances in screening, diagnosis, monitoring, and drug development have resulted in only a 10% increase in 5-year survivorship in the past 20 years. As we lead longer lives, one out of two American men and one in three women will face a cancer diagnosis at some point in their lifetimes.

"It's a disease of life itself and to some extent a disease of aging. There are so many different genes and pathways that can go wrong. It's quite incredible learning about the biology of the disease."

Jay Bitkower, president of ACKC, sees the gift to Dr. Cairns as an exciting chance to make a difference in critical research that will benefit him, as a cancer survivor, and the many others affected each year. "Our members' funding of the PCR system is an opportunity to help expedite Paul Cairns' work to develop a urine screening test for kidney cancer, a therapeutic monitoring capability, and hopefully adding to the scant knowledge of kidney cancer genetics," he says.

ACKC Presents $45,000 check to Paul Cairns.
From left: Patricia Todd, Fred Atkin, Paul Cairns, Ken Youner, and Jay Bitkower.

Download the Fox Chase Press Release (PDF) at the bottom of this page

Paul Cairns Project Summary

Kidney cancer kills, largely because it exhibits no obvious, consistent symptoms and because no diagnostic test is available for detection in the early stages when it is confined to the kidney and can be treated successfully. Because it is a relatively rare form of cancer, it receives short shrift from the government and pharmaceutical companies, the two largest providers of research funding in the U.S.

Yet when kidney cancer is detected early, it can be treated surgically to allow 90% of patients to live healthy, cancer-free lives for five years or more. If kidney cancer migrates to nearby tissue such as a lymph node, the survival rate drops to 60%. If it metastasizes to a distant organ, there is only a 9% chance of surviving five years. If kidney cancer could be detected in a routine lab test, many of the nearly 13,000 American patients it kills each year would stand a good chance of survival.

Paul Cairns, Ph.D., of the Fox Chase Cancer Center in Philadelphia is a pioneering researcher on the trail of genetic markers that identify kidney cancer. He has identified a panel of six genes in all of the different types of kidney cancer tumors that he has studied. In 2003, using his methylation [1] patterning hypothesis in a urine test, he demonstrated an 88% predictive value for kidney cancer - see article inre his early work http://www.hon.ch/News/HSN/516513.html. Based on these preliminary results, the National Cancer Institute's (NCI) Early Detection Research Network awarded Dr. Cairns a $1.2 million 5-year grant to develop an accurate and reliable urine screening test for kidney cancer - see http://www.nci.nih.gov/newscenter/pressreleases/EDRN-BDL.

Dr. Cairns is seeking to acquire a real-time PCR [2] System, which will enable him to greatly increase the efficiency of his research and lead to an increase in the sensitivity of his test. Using this state-of-the-art technology enables researchers to identify one cancer cell in 10,000 normal cells, a tenfold increase in accuracy over the technology that Dr. Cairns used in 2003. His NCI grant does not cover the cost of such a system.

The PCR System will also support Dr. Cairns' other work: developing a quantitative genetic serum test to determine the effectiveness of targeted drug therapies, e.g. Sutent and Sorafenib, in reducing cancer cell population for metastatic cancer patients; researching the genetic differences between indolent and aggressive tumors; uncovering new genetic markers for papillary type kidney cancer, etc.

Action to Cure Kidney Cancer is a grassroots organization of kidney cancer patients and their families that works to raise awareness of kidney cancer. ACKC works with government officials, individuals, and organizations to increase the level of public and private funding for kidney cancer research.

ACKC is greatly encouraged by Dr. Cairns' work and supports his efforts to obtain a real-time PCR system for refining and expediting his research in the development of a urine screening test for kidney cancer. Therefore, we have undertaken to raise the funds Dr. Cairns needs for its purchase. The ACKC Board has voted unanimously to allocate $15,000 (half its assets) as seed money in a project to raise the PCR system's total cost of $45,000.

ACKC is requesting contributions to fund the PCR System, which will support Paul Cairns' work. Donations should be made payable to ACKC targeted for the CAIRNS FUND. Please go to our Donation section at http://www.ackc.org/donate to make your contribution.

For further information, see our extensive interview with Dr. Cairns and an article about his early research.

---

[1] Methylation is a chemical change in DNA that has the effect of turning genes on or off (in this case, turning off tumor suppressor genes).

[2] Polymerase Chain Reaction, a technology for amplifying and making copies of DNA in real-time. The PCR System is manufactured by Applied Biosystems Inc.

Developing a Screening Test for Kidney Cancer: An Interview with Paul Cairns

Paul Cairns, Ph.D., received a five-year $1.2 million grant from NCI in December 2004, as part of the Early Detection Research Network (EDRN). He was interviewed at Fox Chase Cancer Center in Philadelphia on August 31, 2005 by Pat Todd, Ken Youner, and Jay Bitkower of Action to Cure Kidney Cancer (ACKC). A short article describing Dr. Cairns' earlier research on a urine test for kidney cancer can be found on the ACKC Web site. (make live link to the previous story).

ACKC: I went to the NCI Web site, and there's no write-up there for you.

Dr. Cairns: Maybe just the project title is there. You could probably find the abstract if you searched, but the Web site is very painful, as you probably discovered. The project is just started; there will be more as we develop. They're always about a year behind. I haven't finished the first year of the grant yet. Probably they'll put some more information on.

ACKC: Tell us about the grant.

Dr. Cairns: The grant began October 2004 and it's for five years, and if the NCI continues with funding this initiative you can apply for continuation or renewal at the end of the five years. It's obviously competitive. The idea was that although the NCI had the SPORE system (Specialized Programs of Research Excellence) for various tumor types, and stressed translational work, there were still areas where translational work was not moving forward as fast in cancer as the NCI would like. Translational work means from the bench to the clinic. Basic work is basic molecular biology … but how is that applied from clinic to the patient? That's what we call translational, to translate our findings from the lab (bench) to the bedside.

It was felt that more needed to be done in translational cancer research. So the NCI began a number of initiatives, and one of those was the Early Detection Research Network, EDRN), which began in 1998. I missed the first round of funding because I was transitioning from being a post doc at Hopkins to here at Fox Chase, and I hadn't decided exactly what I would work on. However, I attended their meetings because I was interested in early detection. I presented my work in early detection of kidney cancer, and there was a pilot program that you could competitively apply for, so I applied for this small pilot program just to get involved.

I succeeded in getting that grant … it was very small and just for two years … that was only $50,000 a year with no indirect (no money to Fox Chase) … just $50,000 direct cost to the laboratory … each year for two years. That began in July 2002.

Towards the end of 2003 came up the call for another five-year round of funding because there is no mechanism with which to join as a full member of the EDRN. But by getting a pilot project, you can become an associate member; you are allowed to attend their closed meetings. The idea of having closed meetings is that people will not share unpublished data. It's not as bad as it sounds. It's for legitimate reasons that we all move things forward in a collaborative nature, because it's very much a collaboration between clinicians, oncologists, researchers, and even private companies. The NCI has accepted, in today's climate, that if a test is going to move forward fast, you almost certainly need the input from a private company. So they actually encourage you to find private company partners to help move this technology forward. And those private companies can also join as associate members.

When I was about a year through my pilot project, I published the initial study on detection of kidney cancer using methylation as a target in urine. And I also published another paper on profiling gene methylation and a larger series of kidney tumors to make sure it was representative in all grades, stages, cell types, etc.

The first of these studies was in urine, but I did that study quickly. What I then wanted to do was to over-sample for the rarer types of kidney cancer. I had only one or two in the fifty samples. I wanted to make sure I had at least five of some of the rarer types because I wanted to be sure that my detection panel would also pick up the rarer types.

ACKC: What are the rarer types?

Dr. Cairns: Once you get past clear cell, which is 75 to 80%, you've got papillary, which is 10 to 15%. But then you've got oncocytoma (which may or may not be benign), you've got chromophobe carcinoma, you've got collecting duct carcinoma. You've also got 10% (although it's not called renal cell carcinoma) of tumors in the kidney are transitional cell tumors of the renal pelvis, analogous to what we call bladder cancer (the transitional cell epithelia.) Even then there were a few sarcomatoids and undifferentiated renal tumors that people can't assign. I just wanted to make sure and also I wanted to over-sample to get a large much larger number of smaller area tumors. Because I wanted my panel for early detection, I wanted to make it applicable to the entire kidney cancer population. And I wanted in particular to keep going to earlier and earlier tumors to see what the sensitivity is.

ACKC: So you really were making it more applicable to a patient with an abnormal lesion on the kidney?

Dr. Cairns: That's part of the problem. We've never gone to lesions, but with the pressure to publish, to be first, to grant funding, the enemy of the good study is the perfect study. There also is an argument that one should not wait until one has the perfect study anyway because … better to get it out there, so that people can read it, comment on it, feed back, replicate it. You can publish fast or you can publish the perfect study and publish slow.

ACKC: Your work so far has been done on patients with known kidney lesions?

Dr. Cairns: Yes. It may seem counterintuitive if you want to develop an early detection or screening test that eventually you would only have a biopsy or bodily fluids; you would not know if the patient had cancer or not. But the best experimental design is to take patients you are about to get a tumor biopsy from who are about to undergo a nephrectomy and take a pre operative urine sample because then you are able to take whatever you're going to use as a marker for detection, establish it's in the tumor, and you know what to look for in the urine. You can also establish whenever you see something in the urine that is not in the tumor, which would be bad for specificity. So all these early detection studies are designed usually with many samples of a bodily fluid in a tumor from patients about to undergo surgical resection.

Of course, the next step, after you establish that it's widespread in smaller area tumors, is to go to a high-risk population in a blinded study. Not that we necessarily have such other than the VHL (von Hippel-Lindau), which is a unique population. Typically we would go to a high-risk population, where you didn't know they had cancer, but you would do a blinded, maybe retrospective study. And then you might do a blinded prospective study. Essentially within the medical cancer research community, there are guidelines set out on how to move something forward towards clinical trial, and what's acceptable in terms of replicating in other labs doing blinded studies, etc.

ACKC: Can you tell us what methylation is?

Dr. Cairns: Perhaps the biggest advance in the last 20 years in cancer research has been the discovery that cancer is a genetic disease. By that I do not mean that it is inherited … only 5% of cancer is inherited. But sometimes the same genes, whether the mutation is inherited, (whether predisposed to cancer like VHL), or also mutated but in an acquired fashion and the other 95% for "sporadic" cancer acquired by aging or environmental insults with carcinogens during their lifetime. So all cancer is genetic. It makes sense that for these genetic alterations that initiate and drive cancer would be excellent targets for early detection, for prognosis, etc.

The most important class of cancer genes is a class of genes called tumor suppressor genes, which act normally to stop a cell becoming cancerous, as gatekeeper as a checkpoint. They don't promote cancer … they inhibit it. Because they inhibit cancer, they have to be knocked out or lost whereas the other class of cancer genes, oncogenes, are switched on because they promote cancer. We use the traffic light analogy that tumor suppressor genes are the red stop sign and oncogenes are the green go signs. Oncogenes, to get a cancer, they need their expression to be increased, to be revved up. Tumor suppressor genes need their expression to be silenced or knocked out.

All the familial cancer genes, like VHL that have been discovered so far have been tumor suppressor genes. There are technical difficulties in identifying both classes of genes, but oncogenes are a little harder to identify than tumor suppressors. So tumor suppressors appear more common, but that might change in the coming years.

There are three ways to knock out tumor suppressor genes in a cell. One is by deletion whereby you simply lose the gene; literally, a fragment of chromosome is shed from the cell when the cell divides. The other one is point mutation, and that's a sequence change of one of the four bases (from an A to a G; a C to a T). If that occurs in certain spots in the gene, that will render the protein non-functional from functional. The third method is Methylation. That is a chemical change used in the normal instances by the cells to switch genes on and off. Some of our genes don't need to be on all the time during our lifetime or during a cell's lifetime. The cell can mark the promoter of a gene which promotes the expression of a gene to be switched on or off by methylation. The promoter switches on the expression of a gene. The promoter is a region on the gene. Methylation silences the gene. Most of these important genes are protected against methylation in normal cells. But in cancer cells, you see methylation in the promoter regions of these tumor suppressor genes like VHL. Also, when you see the methylation, when you look for expression, the expression of the gene is gone. It's been silenced by the methylation.

In sporadic clear cell, you can see methylation of the VHL gene. Tumor suppressor genes need both alleles … mothers and fathers inactivated. It's not enough. Oncogenes, you need only one copy switched on and you can get cancer. Tumor suppressor genes you have to get both copies knocked out. This is why people who inherit VHL, the incidence of kidney cancer goes up over their lifetime … because they inherit one copy knocked out in all the cells in their body because it's only relevant to their renal cells.

Some time in their lifetime, chances are pretty high just by accident or misfortune, that one of those cells will lose the second copy or have it inactivated. Whereas the rest of us have to lose both copies. Thus, when you look at familial, compared to sporadic renal cancer, you will see a lower age of incidence for the inherited, because they are already one step ahead of the rest of us. Now both alleles have to be knocked out for a cell to become a cancer cell.

You can't inherit methylation in that sense. You can only inherit a point mutation or perhaps a small deletion. You can't inherit a large deletion because it probably would be lethal to the embryo if it was of several genes. So all the inherited mutation from VHL is point mutation. A very small deletion of just a few bases, not the whole gene. You still need the second allele to be knocked out in those people and you need both to be knocked out in people like ourselves. That occurs by any combination of deletion, point mutation, or methylation. Deletion is most common, point mutation next most common, methylation least common. Most commonly you would see deletion of one allele and point mutation of the other. That's most common. But you could see deletion of one allele and methylation of the other. But you could see deletion of both alleles, methylation of both alleles, point mutation next. Similarly in sporadic renal cell. For other genes it's not the least common. For example, the P16, which is involved in renal cancer, in anybody it can be methylated, point mutated, or deleted. We think we know the reason. But basically, it nearly always has two deletions. We think the reason for that is that there is another important gene nearby, and for the price of one hit, or deletion, you can knock out two tumor suppressor genes. The cell favors it, whereas you have to have two independent events, two independent point mutations, or methylations to knock out the gene that way.

ACKC: Why do you choose methylation?

Dr. Cairns: I chose methylation because it has conceptual advantages as a target for early detection. Basically, what you are screening for is always going to be a needle in a haystack in a background of normal cells, in a blood sample, in urine, even perhaps in a needle biopsy. You need something that is technologically quite sensitive. The trouble is, with deletion, you need perhaps 50% of the cells (and therefore the DNA in the specimen you are screening) to be from the cancer, or you won't see it. I can show you an example in the lab and it will be clear to you later. With point mutation, you need 5 to 10% of the cell in your urine, blood, or needle biopsy to be from the tumor, or you'll miss it. It will be masked by the background of all the normal cells that don't have the mutation. With methylation we have a technology that can detect one in 1,000. It is the most sensitive at present with the present technology.

That's the conceptual reason, not just for kidney cancer. There are many other groups using methylation ñ sputum for lung, saliva for head/neck, ductal lavage for breast. It's one of the two most promising technologies in all of cancer detection at the moment. That was my conceptual reason for looking at it. We were one of the first to do so. I tried with deletion and point mutation, but I like methylation for the sensitivity.

The other advantage is, and this is a bit more difficult to explain. For point mutation, you have to know where the point mutation is in the gene to design a probe to go in and search for it. That's great if your cancer gene has a hot spot or mutation. But the VHL doesn't have real hot spots. The mutations are pretty much spread out. Therefore, you would need to have a piece of the tumor to screen for in the fluid. Therefore, you can't really use this for early detection. You can use it for prognosis to see if the tumor's come back or it's all been removed by surgery. But with methylation you don't need to know the status of the gene, whether it's methylated or not in the tumor, to screen for it in the urine. I'll show you an example in the lab.

ACKC: How can you find it in the urine?

Dr. Cairns: It has to be positive in the tumor, but you don't need to know whetherÖ Say I screen myself. I don't just use just one gene. I use a panel of genes because I've previously looked at 100 kidney tumors and established that five or six genes in a hundred kidney tumors … at least one or more of those genes is methylated in 95%. I don't know which one. It might be VHL, it might be P16. But I know if I take my panel of six genes, I've got an approximately 95% chance of picking up methylation that is in that tumor and therefore is in the fluid. I don't need pre-knowledge of the nature of the mutation to design my screening test. I do need it for point mutation. I don't for methylation. You are quite right, I don't know if the VHL gene is methylated or not in a patient with a suspicious mass. But if I look at a large enough panel of genes, I know from my work with 100 or more tumors in the lab that I've got a good chance one of those genes will be methylated in this person's tumor, and therefore I have a target from the sensitivity of my test. "No target" is hopeless. There's also another issue because not all kidney tumors have a gene that's point mutated early on in kidney tumor progression. There's no target in many patients, whereas they do have it in methylated genes.

ACKC: Methylated genes are still considered a mutation?

Dr. Cairns: But remember that cancer is not the disease of one gene. It's a disease of at the minimum five or six genes and maybe a couple of hundred. For example, in prostate cancer, nobody has found a real familial gene yet. No one knows what the earliest genes are in prostate cancer because they haven't found the genes yet. And even one of the candidate genes they have found do not have point mutation. So there is no target to screen for, no matter how sensitive the technology you have. There is no target. But there is a gene, GSTP1, which is methylated in 90% of prostate cancers. Add two or three other genes and you can pick up most of the remaining 10%.

In kidney cancer not all clear cells have VHL mutated. So you are going to miss some. Papillary tumors … none of them has point mutation. So you're not going to be able to screen for those, or chromophobe. So that's one conceptual disadvantage to using VHL point mutation or VHL methylation as a target. You have to use a panel of genes. Cancer, any cancer, not just kidney cancer, is so heterogeneous that you are never going to have a single marker; you have to a panel. But that is no problem with microarray high throughput technology. You've got to think about what you are trying to do, what type of tumor, the nature of the bodily fluid involved, the sensitivity of the technology that you have, and you go with what is best.

ACKC: Clear cell can be methylated also?

Dr. Cairns: Yes. I have to have a panel of genes. Papillary, you can't tell from a scan, you can't tell from a scan what type.

ACKC: In order to have cancer, you have to have more than one gene anyway. So if you found a gene for papillary? Then would you test one gene for papillary and one for clear cell?

Dr. Cairns: It's not as simple as that, because we know VHL is the familial gene for clear cell, probably the only one. We also know that VHL is inactivated in the majority but maybe not all sporadic cancers. It depends how hard you look. The harder you look, the more you pick up.

I think VHL is inactivated in the majority of clear cell, but it hasn't been proven definitively. It's very hard to prove the complete absence of something. We are limited by our technology, time, and money. You go with your best chance first, like you sequence the gene. There's also deep sequencing, you sequence one way across the gene, you sequence the opposite strand because sometimes you pick up things.

ACKC: Will mutation of VHL be sufficient to cause kidney cancer?

Dr. Cairns: No, but it VHL appears to be the initiating event in sporadic. It's a good candidate since it's the familial gene. But it appears to be a very early event. It may not be the earliest event. Familial breast cancer is a completely different disease than sporadic breast cancer. So we don't know.

ACKC: Five or six genes to cause cancer. In clear cell ñ VHL is one of the five or six genes?

Dr. Cairns: Yes. There are others that we don't know about. We don't know definitively. We do know some.

ACKC: Do you know the variations that cause each of the methylation abnormalities?

Dr. Cairns: Methylation itself is the mutation. We don't know all five or six genes. It's far worse than that. Probably there're five or six genes but when each one of those becomes activated or inactivated, there's probably a cascade effect whereby another hundred genes become deregulated as a consequence of that, and probably some of those hundred genes are going to determine whether a certain drug works or not. It's far, far worse. Compared to other cancers, we know very, very little about the genes in kidney cancer after the familial genes because the work hasn't been done. It hasn't been funded or because people haven't wanted to do it.

ACKC: You mention in the e-mail that you are looking at lesions smaller than 3 cm. Why choose that cutoff?

Dr. Cairns: Every few years, the pathologists meet and issue a consensus statement on the pathology of renal cancer. They last met in 1997 and completely changed the size and centimeters of the staging system. They decided you would have T1 (7 cm or under) subdivided into a or b … T1a being 4 cm or under; bigger than 4 cm and less than 7 cm is T1b; over 7 cm but still confined to the kidney is T2. Breaching (peeking out of the kidney) is T3. However, within the community's mindset is this 3 cm cutoff, because a long time ago, decades ago, people used to call things smaller than 3 cm adenomas.

ACKC: You picked methylation because of sensitivity issues?

Dr. Cairns: Yes, and also because it has conceptual advantage for specificity. And it is this: tumor Suppressor genes are not inactivated in normal cells. If you see a cell that looks normal, but has a deletion, point mutation or methylation of VHL or another tumor suppressor gene, I would argue that that cell is not normal, regardless of what it looks like to a pathologist.

If in a cell there is a molecular alteration of a tumor suppressor gene …deletion, point mutation, or methylation … that that cell appears in its morphology and in its size to look normal to a pathologist, I, and many other people, would argue that, nevertheless, that cell is not normal. My ex-boss at Johns Hopkins did a beautiful paper with Hubert Humphrey, the vice president. When Humphrey ran against Nixon, he had some blood in his urine. He went to Hopkins in secret. Five pathologists looked at the cells in the urine. Four said no problem, one said borderline, inflammatory changes, I'm a little concerned. Humphrey heard what he wanted. Eight years later he died of a very invasive bladder cancer. My boss went back (this is when I arrived in America), he had the tumor, the urine samples, and the original samples and he showed the mutation of a tumor suppressor gene, P53, that was in the invasive bladder tumor. And he found it in the urine eight years prior. The widow gave him permission to publish. It went to the New England Journal of Medicine. It clarified molecular medicine for a lot of people.

Remember, cells from early tumors look normal often to a pathologist. At low grade, low stage, they don't look unusual. But nevertheless the conceptual advantage of treating cancer as a genetic disease and using these genetic alterations to screen for is by definition they are the initiating events; they precede obvious cancer. They are there years or decades before cancer gives pain or bleeding symptoms to a patient, before they can be detected by imaging, or before they can be seen on a needle biopsy by a pathologist.

ACKC: What kind of alteration was there in Humphrey's case?

Dr. Cairns: It was a point mutation of P53. P53 is not methylated. It is also infrequently point mutated in kidney cancer. It would be good thing to target, but few renal genes have P53 mutation, unfortunately.

ACKC: What is "serum proteomics"?

Dr. Cairns: In general, the opinion in the field, and my opinion, is that the two currently most promising detection technologies for cancer are methylation and proteomics. Proteomics, the idea is that you use a machine called a mass spectrometer. You don't really go in with a hypothesis of certain genes that you think would be sensitive or specific. You do what is called pattern recognition. You take 50 patients with cancer and 50 patients without cancer. Simply take some of their serum and get the machine to look at the protein spectrum. Then feed the 50 cancer patients to a computer telling them to find some things that are common between all these 50 patients, but not in the other set, the 50 normal. You train the computer without even knowing what you are looking at, like a bar code, to see which spots or proteins are in the cancer patients but never in the normal. Then when you've done the training set, you take another set of 50 people and you test the computer's mathematical algorithm. This is considered very promising because it's a global approach. The computer is doing it. There's no bias allegedly. Certainly within the EDRN, which is the primary organization, methylation and proteomics are regarded as the two most promising.

ACKC: What do you mean by quantitative real time value?

Dr. Cairns: I'll show you examples a little later. What we were doing was a "yes" or "no," which you might think is the perfect way to screen for cancer. But believe it or not, people want the quantitative value … clinicians, companies, etc. We're using that partly because it has other advantages. For some genes we've been able to get down to one in 10,000 because it has a laser which measures fluorescence released from the gene, rather than a dye that binds to the DNA. I'll show you examples and it will all be clear. It also saves us time because we can do it faster.

In real time you actually see the fluorescence readout on computer, you don't have to load a gel to get an answer. Saves us a lot of time and effort. But it also it tells us not whether methylation is there or not, it tells us how much methylation is there.

Now, for early detection, I don't think that's a major advantage, but for monitoring and prognosis, I think it is a major advantage because you have your urine as a base line, you have removal of the tumor (surgical resection), you can check the serum or urine to see if methylation has disappeared or gone. You can monitor if therapy is working by methylation and serum specimen. So it has conceptual advantages for prognosis and monitoring, and it has immediate advantages which I'll show you for time and effort and also to some extent for sensitivity and specificity.

ACKC: Would you define sensitivity and specificity?

Dr. Cairns: Sensitivity is your ability of how many people with cancer you can detect as a positive for your test. Specificity is how many times your test detects a true positive versus a false positive. In general, if you aim for a high sensitivity, your specificity will come down. For example: methylation does not have a high sensitivity for prostate cancer because it's very difficult, although the prostate does secrete into the urethra so it gets into the passage of urine. Not as intuitive as bladder or renal. So the sensitivity is very low. But, because prostate is a common disease, methylation still has use because it's very specific to use in targeted sub populations such as men with high PSA who keep coming back and having a painful needle biopsy that is negative. We don't plan to use it for early detection, but we plan to apply it where the need is.

ACKC: What about for kidney cancer?

Dr. Cairns: For kidney cancer, methylation is sensitive and specific. Even if it wasn't sensitive but was very specific you could add it on to future tests. For bladder cancer you have urinary cytology, blood in the urine. It's not that sensitive or specific. But you can add to it with combination testing.

ACKC: So, specific means a low rate of false positives, and sensitive means a low rate of false negatives?

Dr. Cairns: Yes. But do remember that you don't necessarily need a high sensitivity and a high specificity for something to be useful in cancer. It depends upon the question being asked.

ACKC: Three years ago you had a sensitivity of 88%.

Dr. Cairns: Yes, but bear in mind, those 50 tumors and urines weren't solely stage 1 or stage 2 tumors. I'm biased, but I got a consecutive series of 50 patients' urine. Again, the enemy of a good study is a perfect study

ACKC: Weren't they stage 3 or below?

Dr. Cairns: Yes, I tried to eliminate stage 4. But you could argue that if I then went and did 50 stage 1's, the sensitivity could drop to 60 or 70 with the then current technology. On the other hand, this new machine [PCR System] should get me to one in 10,000 over one in 1,000. We don't know if the urines of a negative are truly cancer cell DNA or if they just have a lower level for the limits of our technology. I suspect that's true. And as the technology improvesÖ

ACKC: So you think the sensitivity was low because of the technology?

Dr. Cairns: Yes. By definition a 1 cm tumor must shed fewer cells into the urine than a 10 cm tumor.

ACKC: So you haven't tested since two years ago?

Dr. Cairns: We're testing all the time. But we haven't, one thing, once you've done a feasibility study is … There are several things we want to do. We want to make sure our panel of markers (and we want to add to it) will pick up all kidney cancers … chromophobes, oncogenes, collecting duct, etc. Because you don't know what the patient has although you can sometimes make a guess on whereabouts on the kidney the lesion is by the X-ray where the lesion is. We want to tweak the technology to make it more sensitive and specific. We want to study smaller and smaller tumors if we can get them. And a major problem here is we don't know the history, the natural progression of kidney cancer.

Colorectal is beautifully mapped out with polyps of certain sizes. But the cysts you see in VHL patients, you do not see in sporadic kidney cancer. We have no idea what comes before a 3 cm kidney cancer tumor.

ACKC: If you can detect a lesion at T3, they're reasonably large, why spend money or time looking for that?

Dr. Cairns: Yes, because you want to detect it when it is a T2 or T1. There are various other things which we want to do which are going to be very tricky and we're going to have to have help from other centers. One is the end stage renal disease. People on dialysis have a higher incidence of renal cancer. We have a chance perhaps, of following some of these patients and getting some very small tumors. Another thing is, autopsy. There was a classic study cited in the fifties in Scandinavia where they took a lot of car crash victims, and showed that some people had incidental renal cancer on autopsy. It hadn't given symptoms, wasn't dangerous, but it was there in autopsies. And this is another critical thing: you never want to design a test like PSA that over-detects people. The critical problem with PSA is that all these people are getting operated on who don't need to be. But you can't differentiate who needs to be and who doesn't. It's a major, major problem. Denmark does the most watchful waiting for prostate cancer, but also has the highest death rate. On the other hand, probably eight out of 10 are having their prostate removed who don't need it.

But you can't differentiate. You don't want over-detection. You actually want to pick a sweet spot, and that sweet spot might be in renal, may be one and a half cm, or from the watchful waiting study, if we divide the watchful waiting tumors into lazy ones … tigers vs. pussycats, that's what they call them in the field.

ACKC: They call them indolent, right?

Dr. Cairns: We have to establish the criteria. Our study will stand or fall on what cuts off. We have to match for size, grade, and stage. You can't take a 7 cm clear cell and say, well, this grew faster than a 3 cm papillary. You have to match for cell type, grade, stage. And then ask why one 7 cm grade 2 clear cell grew faster than another 7 cm.

ACKC: How do you know that?

Dr. Cairns: You have to just do it by following the patient from the moment you get them by X-ray or scans. Then you see it grows faster, but how do you know, where do you look. First of all, you take those two populations and cut them by criteria that we would have to discuss and develop. Then we would basically do global arrays, micro-array screens. We won't assume anything. We're just going to use these microchips with 30,000 genes on and just look for patterns between the two.

ACKC: Like proteomics?

Dr. Cairns: Better than proteomics because it's more established. We don't really know any gene that you could make a conceptual argument that it might beÖ I would scan them first (the likely candidates) from the results that come up from the microarray screen.

ACKC: Assuming these gene defects are there, let's say someone has a nephrectomy, why won't the gene defect still be there?

Dr. Cairns: Sporadic originates from one cell. In familial cancer, because it is inherited in the germ line, it is in every cell in the body. Some tumor suppressor genes that you inherit, you can get cancer in any organ in the body because that suppressor gene is important. But VHL is limited in its tissue specificity to be a risk to kidney cancer. There's a spectrum. And that spectrum is based on the fact that VHL only plays an important role in certain cell types. VHL confers no risk to the breast, to the blood, the lung, etc., but other tumor suppressor genes do.

ACKC: Are you going to follow patients who have had nephrectomies to see what happens to their level of methylation?

Dr. Cairns: We haven't done it with levels, but we've done it with "yes" and "no." We took 25 patients, most from the original 50, who had had a nephrectomy for organ-confined disease, and had no clinical evidence of disease several weeks or months after nephrectomy. We took the postoperative urine specimen and looked to see if the methylation that was in the preoperative urine and the tumor was in the urine or not. Twenty-three of 25 were negative, two were positive (one because we think it is familial, that person had a nephrectomy of one kidney and a partial of the other, so we think it's sitting there even though he has multiple oncocytomas, etc ñ we think he's a Birt-Hogg-DubÈ). The other patient is probably a false positive, but we can't rule out contamination. I'm going to get somebody different in the lab to prepare the second urine in a different room and see what happens. I don't think the test is perfect, but that's what we've written up.

ACKC: Have you found any patients with metastatic disease who have had resections and have seen what their levels are?

Dr. Cairns: No, we haven't, but that's what we intend to do.

ACKC: Want a volunteer?

Dr. Cairns: If we do, the problem is, the rules of the IRB [Ed. note, Institutional Review Board, which approves human subject research protocols] would prevent us from giving you information legally. That's a huge problem for patients. But it has nothing to do with me. The only thing I can tell you is, unless people do these studies, unless people disseminate these studies in publications and talks to the wider community, and unless there's a mechanism for moving this work forward, unless there's a collaboration to move these forward through the NCI and trials, nothing is going to happen. I get letters from people all the time that they've read about the urine test, and it's terrible. There's an argument that perhaps Fox Chase shouldn't do the publicity, but publicity is such an important part of fund raising. I'm not complaining about the letters, but I have to let them down.

ACKC: The genes on the panel, are they all tumor suppressor genes?

Dr. Cairns: In the widest definition, yes. In a narrower definition, no. We're covering the whole field in cancer genetics here. And it's not really possible without delving into the background of each one. For me they have to be methylated in cancer cells and not in normal cells. However, there are arguments, that since aging is related to cancer, sometimes when you look at adjacent normal tissue to, say, a renal tumor, you sometimes find alterations in those cells. But is it contamination? Because you can get little satellite tumors from renal tumors, like you can see a 4-cm but a few 1-mm satellites dotted around that have spread from the main tumor. Or is that those cells are not normal but just look normal to the pathologist? Part of the continuum of cancer in the kidney? Or is it aging? Remember, some of the defects that lead to cancer we acquire through natural aging. So finding somebody's normal kidney cell having a methylation of a gene in a 78-year-old person is perhaps not all that strange.

So the issue's like this. What we are doing, we always look at as large a number as we can of normal people. And also benign or inflammatory people, in my case people with kidney cysts or kidney stones. We try to make sure that the target we are screening for is not in normal people, not in benign or in inflammatory people. Now I have to do a much larger number of those as well to do that. Some genes are methylated at a lower lever in aging normal renal cells. But we hope that, (a), we don't have to use them in the panel, and (b), we'll establish a quantitative cut off with a big enough gap between the level of methylation to tellÖ

ACKC: In the original sample, what did you find?

Dr. Cairns: We did a "yes" or "no" in the original sample. I didn't use those genes. But remember, I only looked at 20 normal and maybe 10 or 12 people with benign kidney disease because again I needed to getÖ.

ACKC: It's possible there can be normal people with methylation?

Dr. Cairns: I don't think so for classical tumor suppressor genes. But I do think that if I were forced to include less classical tumor suppressor genes in my panel, then sensitivity might go up but specificity might go down. But I'm hoping that's not the case.

ACKC: For clear cell cancer, it could be a point mutation but you could also have methylated as well in the tumor.

Dr. Cairns: I'm screening for the methylation. If I take 100 tumors representative of cell type, grade and stage at presentation, and I take a panel of five genes, and I find that 95 or 98 of those tumors have one or more gene methylated, I can be reasonably happy that I have a good chance of having a target to look for in the population at risk for kidney cancer. But I can't be sure. And in fact, we have found a few kidney tumors where none in my panel are methylated. That means that I have to improve the panel, which is another thing we are working on.

ACKC: How do you improve the panel?

Dr. Cairns: I take kidney tumor cell lines that have been in tumors that we grow in tissue culture. Methylation is also a very attractive area to study because it's reversible. Some clinicians think that with drugs, it might even be tested on very terminal kidney patients, because we think it may have some bad side effects as well, technically you could switch a good gene back on by treating a patient (with kidney cancer) with this drug. It has a short half-life. It's been used in some clinical trials. Some very advanced disease kidney cancer patients were eligible for those trials. [Ed. note, this is the demethylating drug, 5-azacytidine].

We treat cultured kidney tumor cells alive in tissue culture with this drug and we monitor them to see that they don't get sick. We take untreated and treated alongside each other with the drug and compare the expression profile on the global microarray. We look at expression, and what the hypothesis is ñ a gene that is methylated is completely switched off. So if we see some genes that have no expression in the untreated but are re-expressed in the treated, that's probably because they are methylated or are they are under the control of a methylated gene. Because the drug demethylates genes so the gene switches back on again. We compare non-treated to drug treated.

We look for expression differences in genes; it's painful. You can't imagine the computer power we need. We're on to 44,000-gene chips now. We then take our best candidates and validate them by looking at the promoter, making sure there was methylation in the untreated, be sure it's methylated in primary tumors because cell lines are always established in advance tumors. They're the only ones that will grow. We go back and look at stage 1, stage 2 clear cell, papillary, and we ask questions. One question we have, for example, is we have to do differential diagnosis, because if you screen somebody's urine, you might pick up a bladder or a prostate. Although I think that's a good thing, clinicians think it's a bad thing. They say you need marker-specific for the kidney.

ACKC: So you find new genes that are methylated and go back to the tumor to see if you can find them there?

Dr. Cairns: First of all, we go back to the tumors to see if they're useful. For example, if VHL is methylated in 20% of clear cell, and we find a gene that's methylated in 80% clear cell, if it's specific as well, it might be useful adding to the panel. If we can find a gene that is only methylated with papillary, like VHL is only methylated in clear cell, even if it's only at a low percentage, it would still be useful for differential diagnosis. There is no conceptual problem with using a panel of 20 or 100 genes in terms of the technology. So we also need to do differential diagnosis, we want to be able to say, is it a kidney or a bladder or a prostate, and in fact we're going to be forced to that as we go forward. And then is it a clear cell or papillary, and is it a tumor or is it an oncocytoma? And then say, is it going to be indolent or aggressive, based on the spectrum of genes that are methylated? Eventually we will want to say, will it respond to this drug or not? Because that's what people are doing in lung and breast cancer. They have tests in breast cancer where if a certain gene is altered, they know the patient will respond or not to drugs. We call this individualized treatment or tailored therapy for patients.

ACKC: So you would treat it with drugs also?

Dr. Cairns: No, but one would hope that drug treatment for renal cancer improves in the future. Drug response is based upon the underlying genetics of all tumors. Whether they respond or not. Now, for lung and breast you have definite individual genes where you can look. You can look for methylation, for point mutation. You can say, this person will or will not respond to the drug. You can better stratify personalized treatment. This is a huge area that's exploding right now.

ACKC: But these genes are diagnostic genes?

Dr. Cairns: No. We're going to choose the genes that methylate early for diagnostic power. But it doesn't mean that there isn't prognostic power even in a stage 1 or stage 2 tumor.

Eventually somebody will hit upon certain genes that dictate whether a particular tumor will or will no respond to a particular drug. This is already happening in lung and breast. You may also be able to predict how fast the tumor will grow, in terms of watchful waiting, by the spectrum of genetic alteration.

ACKC: Have you found other genes there for the panel?

Dr. Cairns: We've found new hypermethylated genes. We're writing up a paper now where we've found three interesting ones. They're all methylated about 40% or more in renal but they're methylated in the different cell types, but we haven't broken it down to whether it would be useful prognostically. We have one other gene which may be methylated in papillary but not clear cell, but we haven't tested enough. All of this took a year to two years. Another thing, because all the kidney cell lines are from very advanced tumors, literally from skin metastases, we started trying to establish and grow our own from the patients here. And that took a long time because we needed so many. We used two propagated tumors in the study. We also wanted to compare genes that were methylated in the two early to the two advanced. There's so much to do.

The interview continues in the cafeteria. The group is discussing a piece of equipment that Dr. Cairns thinks will speed up his research.

ACKC: What is the machine called?

Dr. Cairns: I guess for want of a better name, it's called quantitative real time PCR, but since we do methylation-specific PCR, I would just say real time MSP.

ACKC: Sort of a laser spectrometer?

Dr. Cairns: Basically, every time the DNA is replicated, it releases fluorescence, which is monitored in real time and plotted on a curve. So there is no doing an ordinary PCR, then loading it on a gel, staining it. It's all done in real time.

ACKC: Much faster? More accurate?

Dr. Cairns: Yes. we can recognize 1 in 10,000 under perfect conditions.

ACKC: It costs $93,000 to buy it? Do you have a place for it?

Dr. Cairns: We'll make a place for it. We're going to try to move to a bigger lab. I was fortunate enough to get several grants just before I came up for tenure, so I actually need to hire more people. I really need a lab with seven or eight people for the next five years.

I have no issue with things I want to do. I just need people, money and time to do it. Renal cancer is essentially one of the most neglected cancers. Some very basic things that have been done for other cancers have never been done properly in renal cancer.

ACKC: Could you give us an example?

Dr. Cairns: A simple allele type to see which chromosome workup has never been done properly in kidney cancer, with state-of-the-art technology, and in particular, it's never been done on properly on papillary cancer. All you have are cytogenetic studies from the 1980s, basically. It's unbelievable. Now you have these high density, what I call, gene snip arrays which are the perfect thing, so you really have to do it on some limited number of samples, so some sort of disease like papillary is perfect. But what needs to be done, is you have to collect good specimens, collect blood (normal, renal, tumor) and bank the tumor at minus 80 for the future. You have to have good pathology, good follow-up patient information, and you have to have enough material that you could work on that same tumor for years and share it with other people.

What the field lacks is … people who did a study of a gene in 1980 ran out of those tumor specimens. Now when they want to study a similar gene, they do it in a totally different population. So they can't make comparisons. Because of time pressures, often studies are not designed a well as they could be. What needs to be done for something like kidney cancer is it has to be done, I mean, you go ahead with cutting edge work, but some things also need to be done properly from the beginning with a view to the long term. Because it's never been done. It's not just for kidney cancer. In any cancer, a lot of the populations that laboratories publish are never the same populations. They get depleted and they collect more, but you can never do comparisons, you don't have the follow-up. Like ten years for a patient.

One small thing I want to do is, perhaps with papillary cancer, collect 20 to 30 of the highest quality specimens with all the information, basically throw out anything that I don't have the best information or the best specimen from, and take those and just type those with state of the art technology from the very beginning for gene deletions, gene point mutations, gene arrays and just do it properly. And so I will have that information for the rest of my career, and the same set of patients that I will have follow-up information on, have enough material that I will never run out when I want to do the tenth experiment in five years' time, and I will have enough material to share with other people who have the technological ability that I didn't have.

ACKC: Where can you get 20 specimens?

Dr. Cairns: Well, I have them. I've been banking things for years. I'm just giving you an example. My boss at Hopkins worked on head and neck cancer, and he took note of what the literature was, but it was all over the place. Kidney literature is all over the place. Some people say that VHL is in all; people say P53 is in 10%, some say it is in 40%, but it's all over the place. You can make some judgments based on the lab and the quality of the journal it's published in, but it's difficult without checking it yourself or by using better technology that's come along between now and that study from 12 years ago. I think in many ways there is the argument that just to take something and do it properly. I don't think you have halt the more cutting edge stuff like the methylation detection. But I think you can put somebody in the lab on just starting from first principles but with new better technology we have which gives us more accurate results and just do it. But, as I say, if you're going to do that, you might as well do it properly.

Dr. Cairns: To move the methylation-based detection working urine forward we need to consider optimal urine collection - first void of the morning versus the clinical void - we need to consider how we're going to improve our detection panel of genes, to address differential patterns of diagnoses among prostate, bladder, kidney, and sub-types of kidney, and also to make sure that we have diagnostic coverage of near to 100% as possible of kidney cancers. We need to develop better technology for more sensitivity, specificity, higher throughput, and we think we're going to move toward using quantitative, real-time MSP religiously for every sample. And we also want to contemplate using these quantitative levels and improving the gene panel to address issues of prognosis and monitoring as well and follow-up urine specimens. We're also trying to do marker discovery by doing de-methylation treatments on human kidney cancer cell lines and tissue culture.

ACKC: When do you think you're going to do an update of your original paper, the one with the 88% specificity?

Dr. Cairns: We're doing that now. That's what one of the aims of the grant is. We're stretching ourselves to 3 cm or smaller. It's not that we're getting a nephrectomy of 3 cm every couple of days. We get enough, but it might be a couple of years before we have 100. We also need to get controls. Controls are tricky because you have to go outside Fox Chase for the stones and cysts because they're not seen, because we're just a cancer center. So we have to go to the community hospitals.

ACKC: So it may be a couple of years.

Dr. Cairns: Maybe less, but we're changing the technology so we have to re-optimize the genes in the panel for the fluorescent technology, which we've done for most of them. But nothing's trivial. After that step, which would be a pre-validation study, it would need to be formally validated by the NCI to be shown that it would work in different laboratories.

ACKC: How long would that take?

Dr. Cairns: I can't say. We would hope to achieve all of this within five years.

ACKC: So, basically, we're at least five years away from a commercialized product?

Dr. Cairns: I would think so, but if it's very promising, it could move faster than that.

ACKC: What company makes the machine, the PCR system?

Dr. Cairns: Applied Bio Systems.

ACKC: Is PCR a method everyone can use? Or is it patented?

Dr. Cairns: It's patented, in that you have to pay a certain price to buy the enzyme that makes it work, from Roche, I think. You pay a premium, and there's nothing you can do about that. But they can't stop you adapting the technology.

This is basically what we want to do for real time. We want to see, first of all, if the value of detection in the pre-operative urine, the level, will help determine prognosis, and then we just want to monitor and see real specimens, see disease-free recurrence, partial nephrectomies … has there been something left behind at the margins? … is there spread to the other kidney? Also to monitor for progression of existing disease and response to …

ACKC: Are you starting to do this work?

Dr. Cairns: Yes. We'll publish the follow-up paper with just the "yes" or "no" answer with the 25 patients, and then we'll start trying to do that with the quantitative.

ACKC: When will you do that?

Dr. Cairns: The paper's here somewhere. It's gone to the urologist and come back. I need a sexier title other than "Monitoring tumor suppressor gene methylation in the urine of renal cell patients."

ACKC: This is to predict prognosis?

Dr. Cairns: I need to write the table out better. Cell type, grade, stage. I need to separate these into columns for better presentation: Tumor, pre-op urine, post-op urine, mainly it's methylation, methylation, unmethylated …

Everybody's talked about using methylation for monitoring, but nobody's actually looked. I'm sure they will in the next few months or so. Not much has come out; just a couple of anecdotal reports have appeared.

Fox Chase Cancer Center research shows kidney cancer can be diagnosed in urine

PHILADELPHIA - Laboratory researchers and urologic oncologists from Fox Chase Cancer Center have demonstrated the ability to identify kidney cancer, including localized (stage I) cancer, in the urine of affected patients. The research, supported in part by a grant from the Flight Attendants Medical Research Institute and the National Cancer Institute's Early Detection Research Network, is published in the Dec. 15 issue of the journal Cancer Research.

As with other cancers, an early diagnosis of kidney cancer can result in curative treatment whereas the prognosis for advanced kidney cancer is poor. The challenge in diagnosing cancer early is developing an inexpensive, noninvasive, accurate and simple screening test.

The researchers say a urine test meets these standards. Currently, kidney cancer is diagnosed after radiographic imaging of the kidney, which may include an ultrasound, CT scan and/or MRI. Biopsy of a kidney mass is often difficult to interpret or may give a false negative result and therefore currently confirmation of radiographic results is primarily after surgical excision. There is no protein marker test for kidney cancer as there is for prostate cancer with the PSA test.

"We used a common laboratory procedure to test the urine of 50 patients with kidney cancer," explained Fox Chase molecular biologist Paul Cairns, Ph.D., lead author of the study. "Forty-four of the 50 tests showed gene changes in the urine that were identical to the gene changes found in the tumor samples taken at the time of surgery."

When the same test was conducted on the controls - urine from people without cancer - none showed the relevant gene alterations that were found in the urine from people with cancer.

"The test is remarkably accurate with no false-positives in this study," said Robert G. Uzzo, M.D., a urologic surgeon at Fox Chase and co-author of the paper. "In addition, one of the most impressive outcomes of this research is that the test also identified 27 of the 30 patients with stage I disease. Finding these cancers early means earlier treatment and better prognosis."

The researchers used a molecular DNA-based test called methylation-specific PCR (polymerase chain reaction) to detect genetic alterations that initiate and fuel the onset of cancer. The test searched for six cancer specific tumor-suppressor genes that were altered - causing them to falter in their critical role of preventing errant cell growth. These six genes are usually identified only after a pathologist's review of tumor tissue.

"If these results are confirmed in larger studies, this urine-based test may play a vital role in kidney cancer diagnosis," said Cairns.

###

Fox Chase Cancer Center, one of the nation's first comprehensive cancer centers designated by the National Cancer Institute in 1974, conducts basic, clinical, population and translational research; programs of prevention, detection and treatment of cancer; and community outreach. For more information about Fox Chase activities, visit the Center's web site at www.fccc.edu or call 1-888-FOX CHASE.

Pictures

7900 PCR System

Pat Todd, Jay Bitkower, Ken Youner, Paul Cairns at Fox Chase Cancer Center lab

Postdoc working in Paul Cairns' lab

Attachment	Size
Fox-Chase-Press-Release.pdf	38.13 KB